Back to Models

Kimi k2 thinking

Kimi K2 Thinking is a reasoning variant capable of autonomous long-horizon tasks. It can execute hundreds of sequential tool calls.

Thinking Mode
Parameters
1000000000000 B
Context
256,000 tokens
Released
Jan 7, 2025

Leaderboards

Performance vs. Industry Average

Intelligence

Kimi k2 thinking is of higher intelligence compared to average (4.1), with an intelligence score of 4.3.

Price

Kimi k2 thinking is cheaper compared to average ($1.68 per 1M Tokens) with a price of $0.98 per 1M Tokens.

Latency

Kimi k2 thinking has a higher average latency compared to average (53.05s), with an average latency of 85.59s.

P99 Latency

Kimi k2 thinking has a higher P99 latency compared to average (187.12s), taking 428.50s to receive the first token at P99 (TTFT).

Context Window

Kimi k2 thinking has a smaller context window than average (339k tokens), with a context window of 256k tokens.

Kimi k2 thinking - AutoBench